Glenn E. Morris
Epitope mapping can be used to identify areas of a protein that an antibody recognizes and binds to. Monoclonal antibodies are easier to characterize, but epitope maps can also be produced for poly-clonal antisera.
Key Words: Epitope; conformational; sequential; blotting; phage; biopanning.
Epitope mapping is the process of identifying the region of an antigen that an antibody recognizes (1-3). It is most frequently applied to protein antigens, though methods for mapping carbohydrate antigens are also being developed (4,5). Epitopes on proteins are sometimes a simple sequence of amino acids, although conforma-tional epitopes, formed by bringing together amino acid side chains during protein folding, also are common. Monoclonal antibodies (MAbs) are simpler to map because they recognize a single, unique epitope, whereas polyclonal antisera may recognize several epitopes on the same protein. Nevertheless, it is perfectly possible to map at
From: Methods in Molecular Biology, vol. 295: Immunochemical Protocols, Third Edition.
Edited by: R. Burns © Humana Press Inc., Totowa, NJ
least the major (immunodominant) epitopes recognized by a polyclonal antiserum (6).
The method of choice for epitope mapping depends on a number of factors. The most useful starting point is to establish whether the antibody recognizes a conformational or a sequential epitope. The simplest way to find out whether an epitope is conformational is by Western blotting after sodium dodecyl sulfate polyacrylamide gel electrophoresis. If the antibody still binds after the protein has been boiled in sodium dodecyl sulfate and 2-mercaptoethanol, the epitope is unlikely to be highly conformational. Antibodies against assembled epitopes often display the high avidities and specificities required for immunoassays, but they are difficult to map. Mapping methods for these are rather vague (e.g., competition enzyme-linked immunosorbent assay [ELISA]), incomplete (e.g., antibody protection of antigen from chemical modification), or long and involved (e.g., X-ray studies or in vitro mutagenesis of recombinant antigens). In contrast, sequential epitopes can be mapped by powerful peptide and fragmentation methods. I have discussed in detail the additional considerations involved in choosing an appropriate mapping method (1), such as availability of purified antigen or whether the antigen can be expressed from cloned complimentary deoxyribonucleic acid (cDNA). Details of 30 different mapping methods, mainly for B-cell epitopes, can be found in Epitope Mapping Protocols (1). A more recent volume (7) is especially useful for T-cell epitope mapping.
For the present chapter, I have chosen to present a complete protocol for a phage-displayed random peptide library that is particularly useful for antibodies that work well on Western blots. It has the major advantage that only the antibody and the peptide library are required and, because no antigen is needed, the reader can apply the method to commercially available antibodies or human auto-antibodies. In our experience, some MAbs are not amenable to pep-tide mapping, so we prefer to use a mixture of 5-10 different MAbs (not necessarily against the same antigen) for "biopanning" to ensure a positive result. Selection of phages by biopanning is first carried out using the mixture of antibodies. Colonies of Escherichia coli infected by the selected phage are screened first with the same mixture of MAbs. The positive clones thus identified are then screened with each individual MAb to determine their specificity. MAbs that are able to react with these clones are then removed from the mixture and the biopanning can be repeated with the remaining antibodies ("re-iterative" biopanning; ref. 8). We recommend testing 30-40 clones for MAb binding after the second biopanning because further rounds of panning are often unnecessary and could, in theory, reduce the diversity of peptides selected in favor of higher affinity sequences. For many years, we have used the excellent phage-displayed random peptide libraries developed by George P. Smith of the University of Missouri at Columbia (9-11). Most recently, we have applied this method to panels of MAbs against dystrophin-dystroglycan (12) and the myotonic dystrophy protein kinase (13). There is a rather similar commercial kit, "Ph.D," marketed by New England Biolabs (Beverly, MA), but the Smith method has some important differences that, in our experience, make it much simpler to use. Both methods use random peptide libraries inserted into coat proteins of filamentous phage and these are progressively enriched for epitope-containing phage by biopanning the phage particles against antibody bound to a solid phase. Typically, mouse MAbs are captured onto a small Petri dish coated with rabbit anti-(mouse Ig). After biopanning, the bound phage are expanded by overnight infection of an E. coli suspension culture and recovered from the culture medium for the next round of biopanning. After two or three rounds of biopanning, antibody-positive phage may constitute a high proportion of the total and a cloning step on culture plates is performed in order to identify them. It is at this point that the Smith and the "Ph.D" methods diverge. The Smith method uses phage vectors derived fromfd-tet, a derivative of M13 carrying a tetracycline-resistance marker, while the "Ph.D" method uses M13 itself. M13 colonies in the "Ph.D" method are produced as "turbid plaques" of slow-growing bacteria in a "lawn" of E. coli. These plaques require even more skillful handling than the "clear plaques" produced by lytic phage, such as bacteriophage lambda, and the "Ph.D" protocols recommend an ELISA
method to screen each plaque for antibody-positive phage. This cloning and screening process is greatly simplified in the Smith method by the use offd-tet phages. E. coli infected with these vectors grow as colonies, rather than plaques, because uninfected E. coli fail to grow on tetracycline plates. These colonies can be picked and replicated as master plates on LB-tet agar and also grown on nitrocellulose sheets for screening by a Western blot method. Phage particles released from the growing colonies attach to the nitrocellulose and the bacteria, which might give high backgrounds in a western blot, can be removed by wiping with a sponge.
Another form of displayed peptide library is the FliTrx Random Peptide Display Library (Invitrogen, Paisley, UK), which uses the bacterial flagellum to display random peptide libraries on the E. coli cell surface (14). This library was constructed in the pFliTrx vector, which positions the random peptides in a flagellin (Fli) thioredoxin (Trx) fusion protein. Biopanning with bacteria works surprisingly well in our experience (15) and screening on nitrocellulose is similar to the Smith method.
Phage-displayed 6-mer, 12-mer, or 15-mer peptides are of rather limited use for mapping conformational epitopes, although it is possible to create custom libraries for particular antigens by cloning random fragments of the antigen cDNA into phage (16). These detect conformational epitopes more frequently because the protein fragments expressed are larger. "Mimotopes" (sequences that mimic the shape of the epitope sufficiently to enable antibody binding) can also be isolated from libraries (17) Peptide mimotopes of carbohydrate antigens can even be identified (18) and these may have important applications in vaccine design.
Was this article helpful?